The 3D future

The dawn of the 3D age is upon us. Over a dozen immersive virtual world platforms are competing for our attention — and that’s just in the business sector. Our kids and teenagers are already fully immersed in any of hundreds of different 3D gaming and social worlds.

What will the future look like? Will we have a single dominant proprietary platform, the way that Microsoft rules the desktop? Or will we have an ecosystem of compatible free and proprietary platforms, the way that the Internet runs on both Windows and Apache servers?

The Microsoft path

Microsoft dominated the office and home desktop since the 1980s. Customers were locked in to the platform in several different ways. Switching to another operating system usually meant buying a new computer. Switching also required learning a new interface, and buying all new software and peripherals as well. Finally, switching to another system meant that you were no longer compatible with your friends and colleagues — you could not share documents, for example. Once a sufficiently large group of users were thus locked in, third-party vendors of software and peripherals would have to address this market first, before making products for any other platform. After all, you have to go to where the customers are. And the more products were available for the Microsoft platform, the more people felt that they had to use it.

Is any of this true for virtual worlds today?

Cost of switching. Switching from, say, IBM to Apple requires a significant financial outlay, as well as time and effort. It may also have a social cost, in terms of damaged relationships with friends and colleagues — especially in the early years, when it was very difficult to share files between the two platforms. Today, for an end user, switching from, say, Second Life to ReactionGrid or Blue Mars is free, though the process of setting up an avatar and making new friends could take some time and effort.  For a company, switching from one virtual world to another is more complex. All the buildings and tools created to work in one would would need to be exported, or re-built, to run in the new world. Some virtual worlds use the COLLADA standard for mesh objects, which makes it easier to import and export buildings, but scripts are more difficult to move. Even moving from Second Life to OpenSim can be tricky, since Second Life doesn’t currently offer region exports, and there are some differences in the way that scripts run on the two platforms. In addition, depending on how different the two platforms are, employees may need to undergo additional training to use the new system.

Exclusivity. Ten years ago, a company would either be a Microsoft shop or an Apple shop. Oh, who are we kidding — except for small design firms, they were all Microsoft shops. Having a mixed environment in the workplace was a major hassle. The technology department would need to support two sets of hardware, and two different software platforms. Not to mention all the compatibility issues. Today, however, virtual worlds in the enterprise are used more as single-purpose applications. A training platform, for example. Or a design platform. Or a place for meetings and conferences. As a result, a company can use one virtual world platform for one use case, and another virtual world platform for another use  case. As virtual worlds evolve, their user interfaces will start to standardize, making moving back and forth between platforms easier for users. In addition, most enterprise worlds already support the use of COLLADA mesh objects, and Second Life and OpenSim are expected to do so this year.

Early lead. Microsoft was able to gain an early edge in the platform race by signing a crucial deal with IBM. Today, no single enterprise virtual world platform has that edge. One way to get that edge if a dominant existing platform were to incorporate a virtual world. For example, if the next version of Internet Explorer included a virtual worlds browser that worked together with, say, a Microsoft version of Second Life, then we might see an early lead develop. Another possibility for an early lead was if, say, Google offered a free OpenSim-based virtual world as part of its Google Apps platform and included a virtual worlds browser as part of Chrome. Meanwhile, the race is wide open, with Second Life having the edge in the total number of enterprise end users, and OpenSim having the edge in the total number of private worlds.

The Apache path

Apache — a free, open-source Web server — currently dominates the Internet. Over the past decade, about two thirds of all Websites ran on Apache, twice as many as on Windows. Apache had a couple of years head start over Microsoft. Since it was an open source project, Microsoft couldn’t buy it up, the way it did other potential competitors. In the early days, there were some incompatibilities between the platforms, and some Websites would have a banner saying “best viewed in Netscape” or “best viewed in Microsoft Explorer.” However, these incompatibilities have all but vanished over time, and today any one of a number of browsers — Explorer, Netscape, Firefox, Chrome, and Apple’s Safari can all show Websites running on Apache, Windows Server, Sun, or any other platform.

Free and easy. Apache gained a lot of early adopters because anyone who had a few technical skills could set up a Website and run it out of their own house — or set up a small hosting company and run websites for other people. As a result, we saw an explosion of content coming out from individuals, groups, institutions, and small companies. Large companies weren’t significant customers at the beginning since they had their internal platforms in place for clients and employees to use, usually through direct dial-up connections. The Internet, by comparison, was a security risk, and filled with folks not likely to be spending a lot of money — scientists, students, and hackers. Established platforms like America OnLine were a better fit for marketers because of a high concentration of users.  Today, OpenSim offers a free and relatively easy way of setting up your own virtual world. Individuals are running worlds from their family computers, or setting up small hosting companies to run worlds for others.  Prices for renting an entire OpenSim region from a hosting company start at just $16 a month — in line with early Website hosting costs. Meanwhile, the OpenSim worlds are seen by some as a security risk, with Second Life a better fit for virtual world retail and marketing, and proprietary platforms like Forterra’s Olive a better fit for large enterprises.

Flexible. A Web site running on the Apache system can be viewed through any number of browsers. Similarly, an OpenSim-based virtual world can be viewed through any Second Life-compatible browser, and several independently developed OpenSim browsers, including a Web-based viewer from 3Di. The disassociation between the viewer and the server can cause problems — development has to be slowed down, for example, so that the server doesn’t get ahead of the browser, and vice versa. The servers also have to maintain backwards compatibility. However, this disassociation also means that users have much more choice than they do with a locked-in system. Over time, this means faster evolution of the entire ecosystem, as users choose their preferred viewers and their preferred servers independent of one another. For example, companies can create proprietary alternatives to OpenSim — platforms that are more scalable, or flexible, or serve particular enterprise functions — and still benefit from having access to content, developers, and users familiar with the OpenSim universe.

Interoperable. Back in the early days, before the Internet, users needed special software to access online platforms, and the software was not interchangeable. America OnLine distributed its software on disks and CDs packaged with other products, shipped in magazines, handed out at counters, and mailed directly to people’s homes — and the software could only be used to access their system. To use Compuserve, you needed another disk. At the beginning, you couldn’t even send email or an instant message from one platform to another. Today, we have a similar situation in the world of proprietary virtual worlds. Each requires its own browser or plugin, a separate user account, a new avatar. Avatars can’t travel from one world to another, send messages, or exchange content. In the OpenSim world, by contrast, avatars can travel from one world to another the same way a Web surfer would go from one Website to another. They can even exchange content with users of other grids, or buy products in other worlds and bring them home.

The future

Currently, all indications are that we’re heading for an Apache-style future. The virtual world we’ll be inhabiting five years from now will probably look nothing like the fledging OpenSim builds and Second Life regions that we have today, but will have evolved from them. Non-compatible proprietary platforms will remain in place for special use cases, just as we still have dial-up access for bulletin boards and direct access for corporate data interchange networks. We will get used to seeing certain kinds of content — training, meetings, collaboration — delivered in a virtual environment.

Another five years later, key business functions will begin to migrate into the 3D virtual world, the way they are currently migrating from desktop software to Web apps, and the way they migrated from mainframes to desktops a generation earlier, and from paper to mainframes a generation before that.

So is there any chance for a proprietary, Microsoft-style system to dominate?

Yes, if two things happen simultaneously. First of all, a major distribution channel needs to adopt a proprietary virtual world platform that doesn’t have a compatible open source alternative. This could be Microsoft or another big player purchasing Altadyn or Forterra and then embedding the viewer into their channel.  Second, the platform needs to be accessible enough to encourage mass adoption, and flexible enough to allow enterprise customization and back-end integration.

Can Microsoft or Google do this?

Microsoft is better positioned than Google, since its Internet Explorer dominates the browser market, and it already has a decent share of the Web server market. The company can embed a proprietary viewer in Explorer — or ship a free standalone virtual world browser with its Windows operating system, offer free hosted worlds to individuals, groups, and small businesses, and sell enterprise-class server software to large enterprises.

It won’t be easy. The Second Life and OpenSim worlds are easily edited from within the world itself — users can create new clothing, build new buildings, even raise and lower land. This makes it easy for relative novices with no programming or 3D design skills to create usable environments. To compete, Microsoft will need to build a platform that either has embedded creation tools or a separate easy-to-use design system — not to mention the server and browser software. Currently, no proprietary vendors offers both a platform that supports movement between worlds and in-world creation tools, but it’s possible that Microsoft can buy up a platform that has most of what it needs and build the rest.

Second, Microsoft will run afoul of anti-trust regulations. It already had to face a massive lawsuit over the bundling of its Internet Explorer browser. That doesn’t mean it won’t try, or that some other company won’t try to do something similar. For example, a third-world vendor might sign a deal with Dell or HP to include its software with every shipment of a new computer.

The virtual desktop

And there’s one more possibility. If we assume that eventually, the 3D game-style interface will take over, then we will eventually have a 3D desktop, an operating system built around the 3D environment — a 3D version of Windows. Instead of the desktop metaphor, we would have the house or office building metaphor. We humans are hard-wired to learn to navigate three-dimensional environments. A 3D desktop may be easier to use, and more fun to explore, than the current windows-and-menus approach.

If this is bound to happen eventually anyway, with Apple and Microsoft moving their desktops to 3D,  a smart company may jump ahead of the curve and roll out the platform now.

Such a desktop could itself be a browser — no separate software necessary.

You turn on the computer, and you’re in your virtual house or office. Step out the front door, and the world you’re on stretches in front of you — or private company’s universe, if you’re on an office machine, or, if you’re on a family computer,  your social or residential world of choice. Step through a hypergate — or type a new hypergrid address into your navigation bar — and you’re teleported to another world.

A new upstart can create such a desktop environment and license it to major PC distributors for use by schools, children, and, later, families and companies — or license it to video game vendors to use as the operating platform for their consoles. Nintendo, for example, sold almost 10 million Wii consoles in 2009. If each one included a virtual world browser,  access to your own, editable virtual world region and a free or low-cost virtual world creation system for individuals, groups and small businesses, it would immediately blow Second Life and all other competitors out of the running.

A company that does this successfully will immediately dominate the server market for virtual worlds as well as the browser.  and will also be able to make money from third-party services accessing the platform, the same way that Microsoft now makes money in all these ways from Windows.

And, historically, big companies have had problems transitioning to new platforms. IBM failed to leverage its dominance of the hardware sector into dominance of the operating system — and lost its hardware position as a result. Microsoft was late to the Internet and is an also-ran with Web servers and search engines. PayPal, not existing banks, created a usable online payment system. Amazon dominates online book selling, not a traditional bookstore chain, and eBay owns the online auction business and not an existing auction house.

Given this track record it’s unlikely — though not impossible — that an existing major player will move quickly and aggressively into the 3D space and take it over. It’s slightly more likely that a new player will come out of left field and create a revolutionary new platform.

And the most likely scenario is that OpenSim will follow the Apache development path, creating a 3D future that looks a lot like the Web does today — crazy, mixed-up, insecure, yet supremely appealing.

Maria Korolov